My angel is the manifold

Paul Brzeski
5 min readMay 25, 2017

Naming is the most important decision in a new open source project. It needs to convey a meaning, be ubiquitous, unique, low GI, etc.

I’m coming up to the 5th anniversary of developing Langenium and I’m actually further away from a playable game than the early days of the project.

I made this trailer months into the project… it has more gameplay and functionality than my current build 4.5 years later. It can be a real downer looking at it this way…

What went wrong? This is actually where I expected to be at this point, although I can see how it can look bad without an explanation. In simplistic terms — my idea was too grandiose.

Just building a game is a big undertaking. Today we have many tools and vast communities that can help sidestep the larger time overheads such as pre-built engines, off the shelf art assets, etc. Again, I had bigger plans.

Trying to fill out a franchise of games like Langenium with my own artwork has been a rewarding experience. The research alone has led me to learn new things and I’m happy with how it’s pushed my creative abilities. At the end of the day though, I am trying to build every aspect of this game franchise on my own — writing, coding, modelling, music, marketing, etc.

This week, I realized that the Indie game scene is lacking a certain type of support tool. My new project Manifold is a “graphics program” that will allow anyone to create 2D and 3D artwork, aimed at Indie game developers who need a simplified asset production pipeline.

Step 1. Draw face parts in Illustrator
Step 2. Create sheet of parts for Live2D Cubism Editor
Step 3. Sacrifice countless hours perfecting this mess… srsly omfg.
Step 4. Trick a Time Lord into taking you back in time.. somehow without aging… because you are not going to complete this project in your natural lifetime.

Bias is an important part of developing software like this, I am going to purpose build something for Langenium and let all the other goals hang off of that endgame. Through the rest of this post I’m going to look at other tools I’ve used to develop Langenium over the years and how I need my own new software to create the same results I have now and solve problems I am struggling with — such as scene building.

Manifold came about because I wanted to work towards replacing Illustrator in my asset pipeline. Illustrator is an incredibly complex tool but it’s a bit of a grandpa in today’s world — an overstuffed overbearing monolith of functionality managed by one of the shadier tech companies, Adobe.

No wonder Adobe are world leaders — look at this cutting edge user interface design in 2017. Lasso selection tool??? RECTANGLES!? TRANSPARENT GRADIONTS!?! SHUT UP AND TAKE MEIN MONEH!!

I primarily use Illustrator because I enjoy it’s responsiveness and intuitive layout compared to Inkscape or svgEdit. Illustrator was made in 1987, I was born in 1988. I don’t feel that it’s interface or functionality has meaningfully improved for at least half my lifetime, so it’s been essentially stale for 14.5 years (sorry Adobe… not sorry :D). Making my own flavour of a decent 2D vector editor shouldn’t be too hard.

Inkscape is Illustrator’s unusable Open Source competitor. Just look at that mess.. WTF is going on in that bottom toolbar :( This is why Linux desktop will never happen.

SO 2D vector graphics aside, I also need something to replace Live2D. I’ve tried to work with this interesting solution for 3 years now but I’m frankly over it. The program is good but it just doesn’t mesh well with my single man team approach.

Do you know how big this model is? Good luck loading a city full of these dudes on your phone.

As Live2D is such a configuration/preparation heavy solution, I think it’s wise to come up with my own purpose built approach that lets me somehow draw my characters and prepare them for animation and use in my game engine straight away. If I’m going to aim there, it’d also be nice to go all the way have nice 3D characters.

Sooo.. I need something that looks like this… but I need to create it in a fraction of the development time with an inherently deficient hand drawing ability.. sure!

Artwork for a 3D world is really tricky and I need the ability to reuse things where possible. Using the traditional combo of different forms of art (brief, concept drawing, modelling, painting, optimisation/polish) is so time consuming, a single asset could conceivably kill a project.

In fact, it may be easier to somehow just become immortal and acquire a time machine than trying to create a 3D game by yourself.

Here’s a sampling of some of my work from Langenium over the years:

The penultimate scene that started it all — a lone airship on the water planet with nearby mysterious flying cities and freaky spiked islands in the distance. See that water? That’s a year of my life.
Original Sprite based Jack in an early version of the Helios Station in Epoch Exordium, a culmination of a ton of work using a technique called Constructive Solid Geometry..
The New Jack, built using Illustrator and Live2D. This demo is still available here!
Probably the most effects heavy my ocean scene ever got, the shaders brought my NVIDIA 670GTX to its knees.

There’s so much detail involved in a project like Manifold that I could literally spend days writing this post, but I’ll leave it here. I hope that going through some of the ups and downs I’ve been through over the last near-5 years of trying to build a game series helps shed some light on where I might need to go.. It’s going to be a real challenge controlling the scope and even just putting disparate ideas together.



Paul Brzeski

Sharing my opinion and passions about the many things in life.